Multi-Model Adaptation Learning With Possibilistic Clustering Assumption for EEG-Based Emotion Recognition

被引:0
作者
Dan, Yufang [1 ,2 ]
Tao, Jianwen [1 ]
Zhou, Di [3 ]
机构
[1] Inst Artificial Intelligence Applicat, Ningbo Polytech, Ningbo, Peoples R China
[2] Key Lab Printing Equipment 3D Mfg Coll & Univ Fuji, Fuzhou, Fujian, Peoples R China
[3] Sichuan Univ Arts & Sci, Ind Technol Inst Intelligent Mfg, Dazhou, Peoples R China
关键词
semi-supervised learning; multi-model adaptation; clustering assumption; encephalogram; fuzzy entropy; emotion recognition; FEATURE-EXTRACTION; FEATURE-SELECTION; IMAGES; REGULARIZATION; LOCALIZATION; DIAGNOSIS;
D O I
10.3389/fnins.2022.855421
中图分类号
Q189 [神经科学];
学科分类号
071006 ;
摘要
In machine learning community, graph-based semi-supervised learning (GSSL) approaches have attracted more extensive research due to their elegant mathematical formulation and good performance. However, one of the reasons affecting the performance of the GSSL method is that the training data and test data need to be independently identically distributed (IID); any individual user may show a completely different encephalogram (EEG) data in the same situation. The EEG data may be non-IID. In addition, noise/outlier sensitiveness still exist in GSSL approaches. To these ends, we propose in this paper a novel clustering method based on structure risk minimization model, called multi-model adaptation learning with possibilistic clustering assumption for EEG-based emotion recognition (MA-PCA). It can effectively minimize the influence from the noise/outlier samples based on different EEG-based data distribution in some reproduced kernel Hilbert space. Our main ideas are as follows: (1) reducing the negative impact of noise/outlier patterns through fuzzy entropy regularization, (2) considering the training data and test data are IID and non-IID to obtain a better performance by multi-model adaptation learning, and (3) the algorithm implementation and convergence theorem are also given. A large number of experiments and deep analysis on real DEAP datasets and SEED datasets was carried out. The results show that the MA-PCA method has superior or comparable robustness and generalization performance to EEG-based emotion recognition.
引用
收藏
页数:13
相关论文
共 65 条
[1]  
Atkeson CG, 1997, ARTIF INTELL REV, V11, P11, DOI 10.1023/A:1006559212014
[2]  
Belkin M, 2002, ADV NEUR IN, V14, P585
[3]  
Belkin M, 2006, J MACH LEARN RES, V7, P2399
[4]  
Bishop C. M., 2010, Information Science and Statistics, V8th
[5]   LOCAL LEARNING ALGORITHMS [J].
BOTTOU, L ;
VAPNIK, V .
NEURAL COMPUTATION, 1992, 4 (06) :888-900
[6]   Domain Adaptation Problems: A DASVM Classification Technique and a Circular Validation Strategy [J].
Bruzzone, Lorenzo ;
Marconcini, Mattia .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2010, 32 (05) :770-787
[7]   A Fast, Efficient Domain Adaptation Technique for Cross-Domain Electroencephalography(EEG)-Based Emotion Recognition [J].
Chai, Xin ;
Wang, Qisong ;
Zhao, Yongping ;
Li, Yongqiang ;
Liu, Dan ;
Liu, Xin ;
Bai, Ou .
SENSORS, 2017, 17 (05)
[8]   Unsupervised domain adaptation techniques based on auto-encoder for non-stationary EEG-based emotion recognition [J].
Chai, Xin ;
Wang, Qisong ;
Zhao, Yongping ;
Liu, Xin ;
Bai, Ou ;
Li, Yongqiang .
COMPUTERS IN BIOLOGY AND MEDICINE, 2016, 79 :205-214
[9]  
Chapelle O., 2009, IEEE Trans. Neural Netw., V20, P542
[10]   Selective Transfer Machine for Personalized Facial Expression Analysis [J].
Chu, Wen-Sheng ;
De la Torre, Fernando ;
Cohn, Jeffrey F. .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2017, 39 (03) :529-545